Abstract The lattice preferred orientation (LPO) of both muscovite and biotite were measured by electron backscatter diffraction (EBSD) and these data, together with the LPOs of the other main constituent minerals, were used to produce models of the seismic velocity anisotropy of the Alpine Fault Zone. Numerical experiments examine the effects of varying modal percentages of mica within the fault rocks. These models suggest that when the mica modal proportions approach 20% in quartzofeldspathic mylonites the intrinsic seismic anisotropy of the studied fault zone is dominated by mica, with the direction of the fastest P and S wave velocities strongly dependent on the mica LPOs. The LPOs show that micas produce three distinct patterns within mylonitic fault zones: C-fabric, S-fabric and a composite S–C fabric. The asymmetry of the LPOs can be used as kinematic indicators for the deformation within mylonites. Kinematic data from the micas matches the kinematic interpretation of quartz LPOs and field data. The modelling of velocities and velocity anisotropies from sample LPOs is consistent with geophysical data from the crust under the Southern Alps. The Alpine Fault mylonites and parallel Alpine schists have intrinsic P-wave velocity anisotropies of 12% and S-wave anisotropies of 10%.

Abstract Anhydrite-rich layers within foreland fold and thrust belts are frequently observed to be the weakest horizon of the sequence. Characterizing the microstructure of anhydrite is therefore important for interpreting the larger-scale deformation history of these rocks. Two microstructures from naturally deformed, fine-grained (<15 µm mean grain size) anhydrite samples from the Triassic Evaporites of the Umbria–Marche Apennines, Italy were analysed using electron backscatter diffraction (EBSD). Microstructural observations, misorientation analysis and crystallographic preferred orientation (CPO) determination were carried out on these samples. Both samples have a CPO characterized by alignment of 〈001〉 and distribution of 〈100〉 and 〈010〉 on a great circle normal to this. This anhydrite 〈001〉 ‘fibre texture’ has not been described before. Microstructure A is characterized by a moderate to weak CPO and a weak shape preferred orientation at 55° to 70° from the trace of the 〈001〉 maximum. Low-angle boundaries are revealed by misorientation analysis. A change in grain size from c . 10 to c . 7 µm corresponds to reduction in strength of CPO and reduction in the number of low-angle grain boundaries. Microstructure B is characterized by a very strong CPO. The orientation of the CPO changes between different microstructural domains. The 〈001〉 maximum is always perpendicular to the trace of a strong grain elongation and high-angle grain boundaries have misorientations close to 〈001〉, suggesting that the CPO is geometrically controlled: anhydrite grains are platy with 〈001〉 short axes. The origin of the CPO is therefore unclear but it need not relate to dislocation creep deformation. Whether or not CPO relates to dislocation creep, both samples have a high number of lower-angle grain boundaries and internal grain distortions with 〈010〉 and 〈001〉 misorientation axes. These are indicative of dislocation activity and the data are best explained by slip on either (100)[010] (dominant) and (001)[100] or a combination of these. Neither of these slip systems has been recognized before. Both microstructures are interpreted to have undergone dynamic recrystallization, and the weakening of the CPO with decreasing grain size in microstructure A is suggested to be indicative of a grain-boundary sliding mechanism becoming active. Comparison with experimental data shows that creep mechanisms involving dislocations at the observed grain sizes require the differential stress magnitudes driving deformation to be greater than c . 100 MPa.